<ul data-eligibleForWebStory="true"><li>Researchers from University of Connecticut and University of Minnesota published a paper titled 'TROJAN-GUARD: Hardware Trojans Detection Using GNN in RTL Designs.'</li><li>The paper addresses the increasing risk of hardware trojans (HTs) due to the use of untrusted third-party tools and IPs in chip manufacturing.</li><li>The researchers propose a novel framework utilizing graph neural networks (GNN) tailored for HT detection in large designs like RISC-V cores, with efficient training and inference processes including model quantization.</li><li>The framework achieved a precision of 98.66% and a recall rate of 92.30% in detecting hardware trojans, showcasing its effectiveness and efficiency.</li></ul>

GNN-Based Framework for Hardware Trojan Detection, Including RISC-V Cores

Discover more